If you want to use the huffman.py program, here are some options.

1) Use it directly in a shell

 a) Type   ~mackay/python/compression/huffman/Huffman.py

 b) Type in a list of numbers, one per line

        0.25
        0.5
        0.125
        0.125

 c) Then (on the next line) press CONTROL-d

2) Feed huffman.py with a list of numbers in a file.

 a) First make a list of numbers in a file.

        $    cat > myprobabilities
        0.25
        0.5 
        0.125
        0.125
             
        [Hit Control-d to stop writing to the file]

        [You can make this list of numbers in any way you want,
                using emacs, or a program, for example.]

 b) Use huffman.p thus:
                       
        $  ~mackay/python/compression/huffman/Huffman.py < myprobabilities

 The program will immediately run (no need to press control-d).

3) For your convenience, you may also give names to the different outcomes.
 For example

 a)   $ cat > myprobabilities
         0.25     a    
         0.5      b
         0.125    c
         0.125    d

 b)    $ ~mackay/python/compression/huffman/Huffman.py < myprobabilities

 The output from this example is:

#Symbol Count   Codeword
a       (0.25)  01
b       (0.5)   1
c       (0.12)  000
d       (0.12)  001

4) You can also use this Huffman module from within 
 another python program.